SGM Journals
Research Article

Set of virulence genes and genetic relatedness of O113 : H21 Escherichia coli strains isolated from the animal reservoir and human infections in Brazil

Luis Fernando dos Santos, Kinue Irino, Tânia Mara Ibelli Vaz, Beatriz Ernestina Cabilio Guth

  • 1Departament of Microbiology, Immunology and Parasitology, Universidade Federal de São Paulo, São Paulo, Brazil
  • 2Section of Bacteriology, Adolfo Lutz Institute, São Paulo, Brazil
  • Correspondence
    Beatriz Ernestina Cabilio Guth
    bec.guth{at}unifesp.br

    • Journal of Medical Microbiology 2010; 59(6):634 · https://doi.org/10.1099/jmm.0.015263-0

    View at publisher PubMed

    Abstract

    Escherichia coli strains of serotype O113 : H21 are commonly described as belonging to a Shiga toxin (Stx)-producing E. coli (STEC) pathotype worldwide. Albeit this STEC serotype is frequently identified among cattle and other domestic animals, to the best of our knowledge no human infections associated with STEC O113 : H21 have been registered in Brazil to date. Here, we report the virulence profile and genetic relatedness of a collection of O113 : H21 E. coli strains mainly isolated from the animal reservoir aimed at determining their potential as human pathogens. The strains from the animal reservoir (n=34) were all classified as STEC, whereas the few isolates recovered so far from human diarrhoea (n=3) lacked stx genes. Among the STEC, the stx2d-activatable gene was identified in 85 % of the strains that also carried lpfAO113, iha, saa, ehxA, subAB, astA, cdt-V, espP, espI and epeA; the human strains harboured only lpfAO113, iha and astA. All the strains except one, isolated from cattle, were genetically classified as phylogenetic group B1. High mass plasmids were observed in 25 isolates, but only in the STEC group were these plasmids confirmed as the STEC O113 megaplasmid (pO113). Many closely related subgroups (more than 80 % similarity) were identified by PFGE, with human isolates clustering in a subgroup separate from most of the animal isolates. In conclusion, potentially pathogenic O113 : H21 STEC isolates carrying virulence markers in common with O113 : H21 clones associated with haemolytic uraemic syndrome cases in other regions were demonstrated to occur in the natural reservoir in our settings, and therefore the risk represented by them to public health should be carefully monitored.

    INTRODUCTION

    Among the human diarrhoeagenic Escherichia coli pathotypes, Shiga toxin-producing strains (STEC) are a major cause of infections and have been implicated in important morbidity and mortality worldwide (Griffin & Tauxe, 1991). STEC can produce a spectrum of diseases ranging from mild symptoms to severe clinical outcomes, such as haemolytic uraemic syndrome (HUS) (Paton & Paton, 1998).

    Production of Shiga toxin (Stx) is believed to be the most important event towards HUS development (Karmali et al., 1985). These cytotoxins can be of two main types: Stx1 and Stx2. Members of the Stx1 group are antigenically similar, whereas Stx2 toxins are quite heterogeneous and comprise several variants or subtypes (Paton & Paton, 1998). One of these subtypes, Stx2d, has been further subdivided into two distinct toxins, Stx2d-EH250 (Piérard et al., 1998) and Stx2d-activatable (Melton-Celsa et al., 1996). The designation of the latter toxin derived from the observation that its cytotoxic titres are increased if the toxin is exposed to intestinal mucus prior to cell incubation. Evidence from animal studies also suggested a more virulent behaviour (Lindgren et al., 1993).

    In contrast to infections caused by other pathogenic E. coli, infections due to STEC have a proven zoonotic character (Chapman et al., 2000). These bacteria are largely distributed among ruminant animals, and cattle can be considered the most important natural reservoir. STEC is transmitted from healthy excreting animals to humans via the food chain or by direct contact with the animals or their environment (Caprioli et al., 2005).

    It is known that about 400 different O : H serotypes of E. coli can harbour stx genes, and a considerable proportion of them have already been recovered from human hosts (). Although STEC O157 : H7 is by far the most important serotype in human infections, some non-O157 strains also pose a substantial concern to public health, as they are able to cause illnesses that are comparable in severity to O157-induced diseases (Johnson et al., 2006; Tarr & Neill, 1996). This is the case for STEC O113 : H21, a recognized serotype causing intestinal and extraintestinal infections in many countries (Blanco et al., 2003; Hogan et al., 2001). Although this serotype was mentioned in the first descriptions of STEC pathogens (Karmali et al., 1985), its epidemiological importance remained largely underestimated until the late 1990s, when a small outbreak of HUS occurred in Australia (Paton et al., 1999).

    Normally, O113 : H21 clinical isolates produce Stx2, including the activatable form of this toxin, and lack the eae gene. This gene, which is present on a pathogenicity island termed LEE, is a defining feature of a subset of STEC serogroups such as O157, O111 and O26, known as enterohaemorrhagic Escherichia coli (EHEC). In the absence of eae, other determinants encoded on large virulence plasmids, such as the EHEC haemolysin (Ehx), a subtilase cytotoxin termed SubAB (Paton et al., 2004), and a protein with putative adhesive functions, termed Saa (Paton et al., 2001) (STEC autoagglutinating adhesin), have been implicated in the pathogenesis of HUS.

    E. coli strains of serogroup O113 circulating in Brazil can be related to two distinct diarrhoeagenic pathotypes, enteroaggregative E. coli and STEC (Dos Santos et al., 2007). The latter comprises only strains of serotype O113 : H21, which is one of the most frequent serotypes recovered from cattle in our settings (Cerqueira et al., 1997; Irino et al., 2005). Nonetheless, cases of human infections involving STEC O113 : H21 have not been registered so far in Brazil. However, a few O113 : H21 E. coli (lacking stx genes) strains have been isolated from patients with diarrhoea in our country to date (Vaz et al., 2004; L. F. Dos Santos, unpublished).

    Taking into consideration the importance attributed to the O113 : H21 serotype, especially in the context of STEC infections, the aim of the present study was to further characterize O113 : H21 strains isolated from diverse sources in Brazil, and analyse their genetic relatedness.

    METHODS

    Bacterial strains.

    A total of 37 O113 : H21 E. coli strains isolated during different surveys conducted in several Brazilian states from healthy food-producing animals (dairy or beef cattle n=30, buffalos n=1, goats n=2) (Cerqueira, 2000; Gonzales, 2003; Irino et al., 2005; Leomil et al., 2003; Pigatto et al., 2008; Oliveira et al., 2007, 2008), a meat sample (n=1) (Cerqueira et al., 1997) and human stools (n=3) (Vaz et al., 2004) were studied. These strains had their serotypes, cytotoxic activity and enterohaemolytic phenotype investigated and confirmed by standard procedures (Dos Santos et al., 2007). The only three human strains were isolated from stools of patients with diarrhoea in São Paulo, and lacked stx genes, being herein designated non-STEC. All the other isolates carried stx, and were STEC. Strains were maintained in tryptic soy broth with 15 % glycerol at −70 °C.

    Detection of the stx2d-activatable and stx2d-EH250 genes and other virulence determinants (toxins, putative adhesins and autotransporter proteases).

    To test for the presence of the stx2 activatable allele (stx2d-activatable), a recently described PCR strategy (Zheng et al., 2008) was employed. Gene stx2d-EH250 was investigated according to the PCR-RFLP methodology of Piérard et al. (1998). The presence of several other virulence-associated genes was determined by PCR using the Taq platinum Green PCR mastermix system (Promega). Assays were carried out in total reaction volumes of 25 μl. We sought the sequences related to SubAB, CDT-V, CNF1, CNF2, Cif, EAST-1, KatP, EspP, EspI and EpeA. Cycling conditions and the specific primers employed in the assays were as previously reported (Abe et al., 2008; Beutin et al., 2005; Blanco et al., 1996; Brunder et al., 1999; Cergole-Novella et al., 2007; Leyton et al., 2003; Schmidt et al., 2001). The presence of genes encoding stx1, stx2, ehxA, eae, toxB, saa, lpfAO113 and iha had been previously investigated (Dos Santos et al., 2007).

    Phylogenetic classification.

    Strains were phylogenetically classified as proposed by Clermont et al. (2000).

    Plasmid profiles and confirmation of the STEC O113 megaplasmid (pO113).

    The occurrence of large plasmids was analysed in agarose gels. After extraction of plasmid DNA by the alkaline lysis method (Birnboim & Doly, 1979), plasmid masses were estimated by using the GelCompar II system. E. coli strain R861, with known molecular size plasmid bands, was used as the marker for the analysis. The STEC O113 megaplasmid (pO113) was confirmed by Southern blotting and probing of the extracted plasmid DNA with a peroxidase-labelled DNA fragment of the subAB operon, which was obtained by PCR amplification of the respective genes using primers SubAF and SubBR (Paton et al., 2004).

    PFGE.

    We followed the method described by Gautom (1997), with some modifications. After digestion of the agarose-embedded DNA with XbaI endonuclease (Invitrogen) for 18 h, PFGE was performed on a CHEF-DRIII PFGE apparatus (Bio-Rad) by using an initial pulse time of 5 s and a final pulse time of 35 s over a period of 20 h. Band patterns were analysed by using the GelCompar II system, and the similarity between the restriction patterns was evaluated by using the Dice coefficient similarity (tolerance of 1 %). Strains were considered to have the same PFGE pattern when all bands were identical.

    RESULTS AND DISCUSSION

    Phenotypic characterization of the isolates

    Twenty-four of the 37 strains included in this study had been previously characterized in relation to their serotypes, cytotoxicity to Vero and HeLa cells and expression of EHEC enterohaemolysin (Ehx) (Dos Santos et al., 2007). It was demonstrated that all these strains belonged to the O113 : H21 serotype and that 21 and 14 of them, all from non-human sources, were able to produce cytotoxin and enterohaemolysin, respectively. Thirteen strains were later isolated from more recent surveys, and then added to this study; they also had their O113 and H21 antigens confirmed, and were cytotoxic to Vero and HeLa cells. Moreover, nine of them were enterohaemolytic.

    Search for the stx2d-activatable genotype

    The Stx2d-activatable sequence was identified in 28 (82 %) of the 34 stx-containing strains. Eleven strains (32 %) had this gene as their unique stx allele. The other 17 strains possessed stx2d-activatable in combination with stx1, stx2 and/or stx2c (Table 1⇓). None of the strains possessed stx2d-EH250.

    Table 1.

    Virulence markers and genetic relatedness of O113 : H21 E. coli strains isolated from the animal reservoir and human infections in Brazil

    Screening for the presence of several virulence-associated genes

    The occurrence and distribution of several other virulence markers is also shown in Table 1⇑. The most prevalent gene was lpfAO113, which occurred in all strains, followed by iha, present in 73 % (27 of 37) of the strains. The genes ehxA, subAB, espP and saa were found equally in 62 % of the strains, and the sequences related to epeA, espI, astA and cdt-V could be identified in 59 %, 35 %, 30 % and 19 % of the strains, respectively. The eae, toxB, cnf1, cnf2, cif and katP related sequences were not detected.

    Based on the distribution of the genes investigated, several distinct virulence genotypes or profiles were identified in this study. Among the STEC, the set ehxA, subAB, epeA, espP, lpfAO113, iha and saa, associated or not with cdt-V, was the most prevalent and could be detected in 38 % (13 of 34) of the isolates. The second most prevalent profile was formed by espI, lpfAO113 and iha, associated or not with astA, and was found in 32 % of the stx+ studied strains (11 of 34).

    Plasmid profile characterization

    Twenty-five of the 37 strains harboured at least one large-sized plasmid band compatible with the already described STEC virulence plasmid (>70 MDa). The three non-STEC human isolates also presented high mass plasmid bands, although they were negative by PCR for all the plasmid genes investigated. To confirm that the larger bands visualized in the agarose gels were related to the STEC O113 : H21 megaplasmid (pO113), Southern blot hybridization analysis with a 1.8 kb probe prepared by PCR amplification of the entire subAB operon was performed (Fig. 1⇓). Hybridization was positive in 20 strains, all of them belonging to the STEC group (Table 1⇑). No reactive fragments were observed in two STEC strains and in the three non-STEC human strains, even under low-stringency conditions.

    Figure image not available in archive
    Fig. 1.

    Plasmid profiles and Southern hybridization of O113 : H21 E. coli plasmids with a probe for the subAB operon, a marker for the large plasmid (pO113) of highly virulent STEC strains. M, Plasmid DNA marker (E. coli R861). (a) Agarose gel showing the plasmid profiles of human non-STEC (lanes 1–3) and STEC (lanes 4–11) O113 : H21 strains. (b) Results of the Southern hybridization for the same strains. Negative results were given by the three human isolates and one STEC strain (arrows). The STEC strain giving a negative result lacked large plasmids [lane 5 in (a)] and all the pO113 markers sought by PCR. The band sizes hybridizing with the subAB probe [corresponding to the higher bands visualized in (a)] are as follows: lane 4 ,  121.1 MDa; lane 6, 112 MDa; lane 7, 112 MDa; lane 8, 112 MDa; lane 9, 142.2 MDa; lane 10, 121 MDa; lane 11, 121 MDa.

    Phylogenetic classification and PFGE analysis

    Phylogenetic grouping classified all strains except one in the B1 group. The STEC strain Ec 267/01 isolated from dairy cattle was classified as group D. PFGE analysis allowed the definition of banding patterns in 35 of the 37 isolates. Two strains could not be analysed because of DNA degradation during extraction (Table 1⇑). Twenty-eight PFGE restriction types were identified, forming three distinct clusters that we designated A, B and C (Table 1⇑, Fig. 2⇓). Cluster A was subdivided into six subgroups, A1–A6, and harboured the majority of the strains, including the strains from human sources that were all clustered in subgroup A6. The strains in cluster A showed degrees of similarity that ranged from 78 % to 100 %. Clusters B and C grouped only the animal STEC isolates. In cluster B (composed of 11 strains), four strains presented a single PFGE type (100 % similarity), and two other strains (G397/02 and Ec 727/05) showed more than 90 % similarity with this specific PFGE type. Cluster C was formed by only one strain (Ec 267/01) with the lowest similarity in relation to the others; this strain was the only one phylogenetically classified as D. This isolate was reconfirmed as being O113 by tube agglutination with specific O113 antiserum.

    Figure image not available in archive
    Fig. 2.

    Dendrogram showing the relationships among O113 : H21 STEC and human non-STEC strains based on macrorestriction analysis of genomic DNA with XbaI.

    The further characterization of O113 : H21 STEC strains from several food-producing animals isolated in Brazil showed that they carried virulence profiles that are very similar to the profiles normally presented by highly virulent O113 : H21 STEC strains linked to HUS (Newton et al., 2009). STEC O113 : H21 is considered one of the relevant non-O157 HUS-causing STEC serotypes and belongs to the HUSEC collection proposed by Mellmann et al. (2008). Although non-O157 STEC strains are isolated from diarrhoeal disease and HUS in Brazil (Irino et al., 2007; Souza et al., 2007; Vaz et al., 2004), to the best of our knowledge, human infections related to STEC O113 : H21 have not been described so far in our settings, and the results presently obtained emphasize the importance of ruminant animals as natural reservoirs of potentially pathogenic clones in the context of STEC infections.

    The majority of the strains studied were found to carry stx2d-activatable either alone or in association with stx1. To our knowledge, this is the first description of activatable Stx2 (Stx2d-activatable) in O113 : H21 STEC strains circulating in Brazil. Oliveira et al. (2008) had previously reported the occurrence of this stx subtype in STEC from beef cattle, but those strains belonged to serotypes other than O113 : H21. Previous data have suggested that strains carrying stx2d-activatable are likely to be more pathogenic. A link between Stx2d-activatable-producing strains and severe disease, mostly HUS, has already been demonstrated (Bielaszewska et al., 2006).

    Associations of several virulence genes which are implicated in adhesion and toxin production were found. High frequencies of lpfAO113, iha and saa are in accordance with published data (Tatarczak et al., 2005; Toma et al., 2004). Of particular interest was the observation that most of the strains harboured the novel subtilase toxin gene subAB. Currently there is little information regarding the distribution of subAB (Khaitan et al., 2007), and the data presently obtained reinforce that this operon is widely distributed among O113 : H21 STEC. In addition, almost 19 % of the studied strains, all STEC, were positive for the cdt-V gene, which was previously described as a marker for isolates involved in HUS cases (Bielaszewska et al., 2004). A high prevalence of espP, espI and epeA genes, which belong to the SPATES (serine protease autotransporters in Enterobacteriaceae) family, was found in this study. This is one of the first reports on the occurrence of epeA after its initial description (Leyton et al., 2003). The presence of espI was linked with stx1, stx2d-activatable and astA, and this result differed from previous studies in which this gene was only associated with stx2d-EH250 (Schmidt et al., 2001).

    Sixty-seven per cent (25 of 37) of the strains were found to harbour at least one large plasmid. In the majority of the strains, this element could be characterized as being the pO113. Published observations for human isolates of several STEC serotypes correlated the possession of large plasmids with an enhanced capacity of causing severe disease (Newton et al., 2009). The three non-STEC human isolates also presented high-mass plasmid bands, although they were negative by PCR for all the plasmid genes investigated, suggesting that these plasmids may carry unknown virulence genes.

    PFGE was highly discriminatory, classifying the studied strains in three main groups. The majority of the strains were clustered in group A. However, group B was formed by isolates containing a distinct virulence genotype, which was formed by genes astA and espI always in association with stx1. The absence of certain markers in this profile such as ehxA and subAB suggests that these strains may represent a subcategory of O113 : H21 STEC with intermediate virulence ability.

    The human isolates included in our study were recovered from cases of diarrhoea, suggesting a virulence potential. However, given the fact that we only had three isolates, comparisons and conclusions are more difficult to draw. Although belonging to the O113 : H21 serotype, which is common among STEC isolates, they lacked stx-related genes upon isolation and presented a very limited set of virulence markers. Moreover, they were not genetically related to STEC strains, as indicated by plasmid profile and PFGE analysis. The possibility that they have lost stx genetic sequences exists (Karch et al., 1992). Additional phylogenetic and in vivo studies are ongoing so that it will be possible to determine whether such strains are truly pathogenic and their relationships with the STEC pathotype. Lastly, the risk represented by animal O113 : H21 STEC isolates to public health should be carefully monitored as such strains are widely present in the natural reservoir in our settings as in many other countries.

    Acknowledgments

    The authors gratefully acknowledge the assistance of Sylvia P. Cardoso Leão for dendrogram analysis. This study was supported by grants from Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), process number 05/04634-8, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - Brasília) and Programa de Apoio a Núcleos de Excelência PRONEX MCT/CNPq/FAPERJ. L. F. dos S. receives a research fellowship from FAPESP, process number 06/60628-0.

    References

    1. Abe, C. M., Salvador, F. A., Falsetti, I. N., Vieira, M. A. M., Blanco, J., Blanco, J. E., Blanco, M., Machado, A. M. O., Elias, W. P. & other authors (2008). Uropathogenic Escherichia coli (UPEC) strains may carry virulence properties of diarrheogenic E. coli. FEMS Immunol Med Microbiol 52, 397–406.
    2. Beutin, L., Kaulfuss, S., Herold, S., Oswald, E. & Schmidt, H. (2005). Genetic analysis of enteropathogenic and enterohemorrhagic Escherichia coli serogroup O103 strains by molecular typing of virulence and housekeeping genes and pulsed-field gel electrophoresis. J Clin Microbiol 43, 1552–1563.
    3. Bielaszewska, M., Fell, M., Greune, L., Prager, R., Fruth, A., Tschäpe, H., Schmidt, M. A. & Karch, H. (2004). Characterization of cytolethal distending toxin genes and expression in shiga toxin-producing Escherichia coli strains of non-O157 serogroups. Infect Immun 72, 1812–1816.
    4. Bielaszewska, M., Friedrich, A. W., Aldick, T., Schurk-Bulgrin, R. & Karch, H. (2006). Shiga toxin activatable by intestinal mucus in Escherichia coli isolated from humans: predictor for a severe clinical outcome. Clin Infect Dis 43, 1160–1167.
    5. Birnboim, H. C. & Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7, 1513–1523.
    6. Blanco, M., Blanco, J. E., Blanco, J., Alonso, M. P., Balsalobre, C., Mourino, M., Madrid, C. & Juarez, A. (1996). Polymerase chain reaction for detection of Escherichia coli strains producing cytotoxic necrotizing factor type 1 and type 2 (CNF1 and CNF2). J Microbiol Methods 26, 95–101.
    7. Blanco, J., Blanco, M., Blanco, J. E., Mora, A., Gonzáles, E. A., Bernárdez, M. E., Alonso, M. P., Coira, A., Rodriguez, A. & other authors (2003). Verotoxin-producing Escherichia coli in Spain: prevalence, serotypes, and virulence genes of O157 : H7 and non-O157 VTEC in ruminants, raw beef products, and humans. Exp Biol Med (Maywood) 228, 345–351.
    8. Brunder, W., Schmidt, H., Frosch, M. & Karch, H. (1999). The large plasmids of Shiga-toxin-producing Escherichia coli (STEC) are highly variable genetic elements. Microbiology 145, 1005–1014.
    9. Caprioli, A., Morabito, S., Brugere, H. & Oswald, E. (2005). Enterohemorrhagic Escherichia coli: emerging issues on virulence and modes of transmission. Vet Res 36, 289–311.
    10. Cergole-Novella, M. C., Nishimura, L. S., dos Santos, L. F., Irino, K., Vaz, T. M. I., Bergamini, A. M. & Guth, B. E. C. (2007). Distribution of virulence profiles related to new toxins and putative adhesins in Shiga toxin-producing Escherichia coli isolated from diverse sources in Brazil. FEMS Microbiol Lett 274, 329–334.
    11. Cerqueira, A. M. F. (2000). Shiga toxin-producing Escherichia coli (STEC): prevalence in animal reservoir, virulence markers and analysis of clonal structure. PhD thesis, Universidade Federal de São Paulo, Escola Paulista de Medicina, São Paulo, Brazil.
    12. Cerqueira, A. M. F., Tibana, A. & Guth, B. E. C. (1997). High occurrence of Shiga-like toxin-producing strains among diarrheagenic Escherichia coli isolated from raw beef products in Rio de Janeiro City, Brazil. J Food Prot 60, 177–180.
    13. Chapman, P. A. C. A., Cerdan Malon, A. T. M. S. & Harkin, M. A. (2000). A one year study of Escherichia coli O157 in raw beef and lamb products. Epidemiol Infect 124, 207–213.
    14. Clermont, O., Bonacorsi, S. & Bingen, E. (2000). Rapid and simple determination of the Escherichia coli phylogenetic group. Appl Environ Microbiol 66, 4555–4558.
    15. Dos Santos, L. F., Gonçalves, E. M., Vaz, T. M. I., Irino, K. & Guth, B. E. C. (2007). Distinct pathotypes of O113 Escherichia coli strains isolated from humans and animals in Brazil. J Clin Microbiol 45, 2028–2030.
    16. Gautom, R. K. (1997). Rapid pulsed-field gel electrophoresis protocol for typing of Escherichia coli O157 : H7 and other gram-negative organisms in one day. J Clin Microbiol 35, 2977–2980.
    17. Gonzales, A. G. M. (2003). Phenotypic and genotypic characteristics of Shiga toxin-producing Escherichia coli (STEC) isolated from cattle in Rio de Janeiro State. PhD thesis, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil.
    18. Griffin, P. M. & Tauxe, R. V. (1991). The epidemiology of infections caused by Escherichia coli O157 : H7, other enterohemorrhagic E. coli, and the associated hemolytic uremic syndrome. Epidemiol Rev 13, 60–98.
    19. Hogan, M. C., Gloor, J. M., Uhl, J. R., Cockerill, F. R. & Milliner, D. S. (2001). Two cases of non-O157 Escherichia coli hemolytic uremic syndrome caused by urinary tract infection. Am J Kidney Dis 38, E22
    20. Irino, K., Kato, M. A., Vaz, T. M. I., Ramos, I. I., Souza, M. A., Cruz, A. S., Gomes, T. A. T., Vieira, M. A. M. & Guth, B. E. C. (2005). Serotypes and virulence markers of Shiga toxin-producing Escherichia coli (STEC) isolated from dairy cattle in São Paulo State, Brazil. Vet Microbiol 105, 29–36.
    21. Irino, K., Vaz, T. M. I., Medeiros, M. I., Kato, M. A. M. F., Gomes, T. A. T., Vieira, M. A. M. & Guth, B. E. C. (2007). Serotype diversity as a drawback in the surveillance of Shiga toxin-producing Escherichia coli infections in Brazil. J Med Microbiol 56, 565–567.
    22. Johnson, K. E., Thorpe, C. M. & Sears, C. L. (2006). The emerging clinical importance of non-O157 Shiga toxin-producing Escherichia coli. Clin Infect Dis 43, 1587–1595.
    23. Karch, H., Meyer, T., Russmann, H. & Heesemann, J. (1992). Frequent loss of Shiga-like toxin genes in clinical isolates of Escherichia coli upon subcultivation. Infect Immun 60, 3464–3467.
    24. Karmali, M. A., Petric, M., Lim, C., Fleming, P. C., Arbus, G. S. & Lior, H. (1985). The association between idiopathic hemolytic uremic syndrome and infection by verotoxin-producing Escherichia coli. J Infect Dis 151, 775–782.
    25. Khaitan, A., Jandhyala, D. M., Thorpe, C. M., Ritchie, J. M. & Paton, A. W. (2007). The operon encoding SubAB, a novel cytotoxin, is present in Shiga toxin-producing Escherichia coli isolates from the United States. J Clin Microbiol 45, 1374–1375.
    26. Leomil, L., Aidar-Ugrinovich, L., Guth, B. E. C., Irino, K., Vettorato, M. P., Onuma, D. L. & Castro, A. F. P. (2003). Frequency of Shiga toxin-producing Escherichia coli (STEC) isolates among diarrheic and non-diarrheic calves in Brazil. Vet Microbiol 97, 103–109.
    27. Leyton, D. L., Sloan, J., Hill, R. E., Doughty, S. & Hartland, E. L. (2003). Transfer region of pO113 from enterohemorrhagic Escherichia coli: similarity with R64 and identification of a novel plasmid-encoded autotransporter, EpeA. Infect Immun 71, 6307–6319.
    28. Lindgren, S. W., Melton, A. R. & O'Brien, A. D. (1993). Virulence of enterohemorrhagic Escherichia coli O91 : H21 clinical isolates in an orally infected mouse model. Infect Immun 61, 3832–3842.
    29. Mellmann, A., Bielaszewska, M., Köck, R., Friedrich, A. W., Fruth, A., Middendorf, B., Harmsen, D., Schmidt, M. A. & Karch, H. (2008). Analysis of collection of hemolytic uremic syndrome-associated enterohemorrhagic Escherichia coli. Emerg Infect Dis 14, 1287–1290.
    30. Melton-Celsa, A. R., Darnell, S. C. & O'Brien, A. D. (1996). Activation of Shiga-like toxins by mouse and human intestinal mucus correlates with virulence of enterohemorrhagic Escherichia coli O91 : H21 isolates in orally infected, streptomycin-treated mice. Infect Immun 64, 1569–1576.
    31. Newton, H. J., Sloan, J., Bulach, D. M., Seemann, T., Allison, C. C., Tauschek, M., Robins-Browne, R. M., Paton, J. C., Whittam, T. S. & other authors (2009). Shiga toxin-producing Escherichia coli strains negative for locus of enterocyte effacement. Emerg Infect Dis 15, 372–380.
    32. Oliveira, M. G., Brito, J. R., Carvalho, R. R., Guth, B. E. C., Gomes, T. A. T., Vieira, M. A. M., Kato, M. A., Ramos, I. I., Vaz, T. M. I. & Irino, K. (2007). Water buffaloes (Bubalus bubalis) identified as an important reservoir of Shiga toxin-producing Escherichia coli in Brazil. Appl Environ Microbiol 73, 5945–5948.
    33. Oliveira, M. G., Brito, J. R., Gomes, T. A. T., Guth, B. E. C., Vieira, M. A. M., Naves, Z. V., Vaz, T. M. I. & Irino, K. (2008). Diversity of virulence profiles of Shiga toxin-producing Escherichia coli serotypes in food-producing animals in Brazil. Int J Food Microbiol 127, 139–146.
    34. Paton, J. C. & Paton, A. W. (1998). Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin Microbiol Rev 11, 450–479.
    35. Paton, A. W., Woodrow, M. C., Doyle, R. M., Lanser, J. A. & Paton, J. C. (1999). Molecular characterization of a Shiga toxigenic Escherichia coli O113 : H21 strain lacking eae responsible for a cluster of cases of hemolytic-uremic syndrome. J Clin Microbiol 37, 3357–3361.
    36. Paton, A. W., Srimanote, P., Woodrow, M. C. & Paton, J. C. (2001). Characterization of Saa, a novel autoagglutinating adhesin produced by locus of enterocyte effacement-negative Shiga-toxigenic Escherichia coli strains that are virulent for humans. Infect Immun 69, 6999–7009.
    37. Paton, A. W., Srimanote, P., Talbot, U. M., Wang, H. & Paton, J. C. (2004). A new family of potent AB5 cytotoxins produced by Shiga toxigenic Escherichia coli. J Exp Med 200, 35–46.
    38. Piérard, D., Muyldermans, G., Moriau, L., Stevens, D. & Lawers, S. (1998). Identification of new verocytotoxin type 2 variant B-subunit genes in human and animal Escherichia coli isolates. J Clin Microbiol 36, 3317–3322.
    39. Pigatto, C. P., Schocken-Iturrino, R. P., Souza, E. M., Pedrosa, F. O., Comarella, L., Irino, K., Kato, M. A., Farah, S. M., Warth, J. F. & Fadel-Picheth, C. M. (2008). Virulence properties and antimicrobial susceptibility of Shiga toxin-producing Escherichia coli strains isolated from healthy cattle from Paraná State, Brazil. Can J Microbiol 54, 588–593.
    40. Schmidt, H., Zhang, W. L., Hemmrich, U., Jelacic, S., Brunder, W., Tarr, P. I., Dobrindt, U., Hacker, J. & Karch, H. (2001). Identification and characterization of a novel genomic island integrated at selC in locus of enterocyte effacement-negative, Shiga toxin-producing Escherichia coli. Infect Immun 69, 6863–6873.
    41. Souza, R. L., Nishimura, L. S. & Guth, B. E. C. (2007). Uncommon Shiga toxin-producing Escherichia coli serotype O165 : HNM as cause of hemolytic uremic syndrome in São Paulo, Brazil. Diagn Microbiol Infect Dis 59, 223–225.
    42. Tarr, P. I. & Neill, M. A. (1996). The problem of non-O157 : H7 Shiga toxin (verocytotoxin) producing Escherichia coli. J Infect Dis 174, 1136–1139.
    43. Tatarczak, M., Wieczorek, K., Posse, B. & Osek, J. (2005). Identification of putative adhesin genes in shigatoxigenic Escherichia coli isolated from different sources. Vet Microbiol 110, 77–85.
    44. Toma, C., Martinez-Espinosa, E., Song, T., Miliwebsky, E., Chinen, I., Iyoda, S., Iwanaga, M. & Rivas, M. (2004). Distribution of putative adhesins in different seropathotypes of Shiga toxin-producing Escherichia coli. J Clin Microbiol 42, 4937–4946.
    45. Vaz, T. M. I., Irino, K., Kato, M. A. M. F., Dias, A. M. G., Gomes, T. A. T., Medeiros, M. I. C., Rocha, M. M. M. & Guth, B. E. C. (2004). Virulence properties and characteristics of Shiga toxin-producing Escherichia coli strains in São Paulo, Brazil, from 1976 through 1999. J Clin Microbiol 42, 903–905.
    46. Zheng, J., Cui, S., Teel, L. D., Zhao, S., Singh, R., O'Brien, A. D. & Meng, J. (2008). Identification and characterization of Shiga toxin type 2 variants in Escherichia coli isolates from animals, food, and humans. Appl Environ Microbiol 74, 5645–5652.